bioRxiv preprint doi: https://doi.org/10.1101/2020.07.29.227710; this version posted July 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Three-Component Model of the Spinal Nerve Branching Pattern, based on the View of the Lateral Somitic Frontier and Experimental Validation Shunsaku Homma1*, Takako Shimada1, Ikuo Wada2, Katsuji Kumaki3, Noboru Sato3, and Hiroyuki Yaginuma1 1 Department of Neuroanatomy and Embryology, 2 Department of Cell Science, Institute of Biomedical Sciences, Fukushima Medical University, 1 Hikarigaoka, Fukushima, 960-1295 JAPAN 3 Division of Gross Anatomy and Morphogenesis, Niigata University Graduate School of Medical and Dental Sciences, Niigata, 951-8510 JAPAN * Corresponding author: S. Homma, [email protected]. bioRxiv preprint doi: https://doi.org/10.1101/2020.07.29.227710; this version posted July 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. ABSTRACT One of the decisive questions about human gross anatomy is unmatching the adult branching pattern of the spinal nerve to the embryonic lineages of the peripheral target muscles. The two principal branches in the adult anatomy, the dorsal and ventral rami of the spinal nerve, innervate the intrinsic back muscles (epaxial muscles), as well as the body wall and appendicular muscles (hypaxial muscles), respectively. However, progenitors from the dorsomedial myotome develop into the back and proximal body wall muscles (primaxial muscles) within the sclerotome-derived connective tissue environment. In contrast, those from the ventrolateral myotome develop into the distal body wall and appendicular muscles (abaxial muscles) within the lateral plate-derived connective tissue environment. Thus, the ventral rami innervate muscles that belong to two different embryonic compartments. Because strict correspondence between an embryonic compartment and its cognate innervation is a way to secure the development of functional neuronal circuits, this mismatch indicates that we may need to reconcile our current understanding of the branching pattern of the spinal nerve with regard to embryonic compartments. Accordingly, we first built a model for the branching pattern of the spinal nerve, based on the primaxial-abaxial distinction, and then validated it using mouse embryos. In our model, we hypothesized the following: 1) a single spinal nerve consists of three nerve components: primaxial compartment-responsible branches, a homologous branch to the canonical intercostal nerve bound for innervation to the abaxial compartment in the ventral body wall, and a novel class of nerves that travel along the lateral cutaneous branch to the appendicles; 2) the three nerve components are discrete only during early embryonic periods but are later modified into the elaborate adult morphology; and 3) each of the three components has its own unique morphology regarding trajectory and innervation targets. Notably, the primaxial compartment-responsible branches from the bioRxiv preprint doi: https://doi.org/10.1101/2020.07.29.227710; this version posted July 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. ventral rami have the same features as the dorsal rami. Under the above assumptions, our model comprehensively describes the logic for innervation patterns when facing the intricate anatomy of the spinal nerve in the human body. In transparent whole-mount specimens of embryonic mouse thoraces, the single thoracic spinal nerve in early developmental periods trifurcated into superficial, deep, and lateral cutaneous branches; however, it later resembled the adult branching pattern by contracting the superficial branch. The superficial branches remained segmental while the other two branches were free from axial restriction. Injection of a tracer into the superficial branches of the intercostal nerve labeled Lhx3-positive motoneurons in the medial portion of the medial motor column (MMCm). However, the injection into the deep branches resulted in retrograde labeling of motoneurons that expressed Oct6 in the lateral portion of the medial motor column (MMCl). Collectively, these observations on the embryonic intercostal nerve support our model that the spinal nerve consists of three distinctive components. We believe that our model provides a framework to conceptualize the innervation pattern of the spinal nerve based on the distinction of embryonic mesoderm compartments. Because such information about the spinal nerves is essential, we further anticipate that our model will provide new insights into a broad range of research fields, from basic to clinical sciences. bioRxiv preprint doi: https://doi.org/10.1101/2020.07.29.227710; this version posted July 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. INTRODUCTION The peripheral branching pattern of the spinal nerve constitutes an essential chapter of the human gross anatomy, and has long been a subject of study in basic and clinical medicine. Nevertheless, a fundamental puzzle has remained: the branching pattern in the adult anatomy does not reflect the embryonic lineages of the peripheral target muscles. In the adult, once it leaves the vertebrae, the spinal nerve bifurcates into the dorsal and ventral rami. The dorsal ramus innervates the back muscles, whereas the ventral ramus innervates the body wall and appendicular muscles (Standring, 2015). The terms of 'epaxial' and 'hypaxial' have been used to describe each respective target of motor innervations by the dorsal and ventral rami (Spörle, 2001). However, the 'epaxial' and 'hypaxial' distinction does not correspond to embryonic somitic lineages (Burke and Nowicki, 2003). The medial portion of the somite gives rise to the back muscles and some body wall muscles, such as the proximal intercostal muscles, whereas the lateral portion gives rise to the distal body wall muscles and appendicular muscles (Ordahl and Le Douarin, 1992; Olivera-Martinez et al., 2000). Two muscle lineages develop exclusively in different connective tissue environments, derived either from the sclerotome or from the lateral plate mesoderm (LPM). The term 'primaxial' is used to describe the muscles that are derived from the medial somite and are differentiated in the sclerotome-derived connective tissue environment, whereas 'abaxial' for muscles from the lateral somite and in the LPM-derived connective tissue environment (Burke and Nowicki et al., 2003). The boundary between the primaxial and abaxial muscle-containing compartments is named the lateral somitic frontier (Nowicki et al., 2003; Burke and Nowicki, 2003; Durland et al., 2008). Thus, the ventral ramus innervates muscles located before and beyond the lateral somitic frontier, whereas the dorsal ramus of the spinal nerve innervates muscles only before the frontier. Hence, the dorsal and ventral rami of the spinal nerve in adult anatomy bioRxiv preprint doi: https://doi.org/10.1101/2020.07.29.227710; this version posted July 30, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. is thought of as a secondary modification on an initial differential innervation laid down during early developmental periods (Gilbert and Barresi, 2016). However, an original innervation pattern based on the primaxial-abaxial distinction has not yet been investigated, and its development into the adult morphology has yet to be analyzed. As the muscles have two different lineages, the target connective tissue of sensory innervation, the dermis, also has two separate embryonic origins (Mauger, 1972; Durland et al., 2008; Fliniaux et al., 2004). The dermis in the primaxial and abaxial compartments are from the somite and the LPM, respectively. Dual sources of the dermis inevitably bring us the second related puzzle of how a single sensory dermatome in the body wall receives coherent sensory innervations. The axial specification by the Hox gene-codes is different between the paraxial and lateral plate mesoderm (Cohn et al., 1997; Nowicki and Burk, 2000). However, current dermatome maps employ the axial levels of paraxial somites, from which the dermis is derived, even for the abaxial compartments. The incompatible axial specification of the dermis and spinal nerve brings into puzzle the global alignment of the spinal nerve with the lateral structures. The third puzzle is how the columnar organization of spinal motoneurons reflects embryonic innervation patterns to muscles in the primaxial and abaxial compartments. The motoneurons in the spinal cord are organized into two primary columns, the medial motor column (MMC) and lateral motor column (LMC). The MMC is further subdivided into medial (MMCm) and lateral (MMCl) portions (Callister et al., 1987; Gutman et al., 1993; Kitamura and Richmond, 1994; Vanderhorst and Holstege, 1997; Watson, 2009). MMCm motoneurons innervate the back muscles through the dorsal rami, whereas the MMCl and LMC motoneurons innervate the body wall and appendicular muscles through the ventral rami. This columnar distinction was initially prescribed based on the locations of a